robsphdfandomcom-20200215-history
Literature review
'The time course of visual encoding. ' ' ' The time course of visual perception. Using ERP measures, occipital activation occurs 90-100 ms after stimulus onset, which is believed to represent low level feature processing. (Madec et al., 2012, Tarkiainen, Cornelissen, & Salmelin, 2002; Tarkiainen, et al., 1999). However no studies I have been able to find compare the processing times between or within low level features. Within 100-150 ms the viewer is able to get the ‘gist’ of a visual scene or image. In other words, they have an abstract or symbolic idea of what they are viewing. For example, they know that they are viewing an indoor or an outdoor scene, but would not be able to describe the details of the scene. At this stage they have not yet created a conscious visual perception of the image (Liu et al., 2009). This early activation is then followed by fronto-temporal activation at around 170 ms after stimulus onset, which is believed to correspond to stimulus identification (Madec, et al., 2012). Further activation in occipital regions at around 220 ms after stimulus onset is believed to correspond to the feed-back ‘hypothesis testing’ (Madec et al., 2012). Further processing may be required for additional semantic knowledge, conscious perception, or encoding to memory (Hedge 2002). These studies highlight differences between low level feature processing, object identification, and encoding to memory; however they do not focus on differences in encoding time for different dimensions of a feature, such as different colours. The prioritisation of different features in attention selection and memory encoding. '' The selection of colour occurs before the selection of other identifying features, such as size, shape or orientation (Eimer, 1996; karayanidis & Michie, 1997). Colour was attended to when the task was to attend to pattern, but not vice versa (Michie et al., 1999). Therefore, based on the literature, it may be that colour has a special status in attentional selection, when it is linked to other non-spatial features, such as shape (Proverbio et al., 2004). Tatlet, Gilchrist, and Rusted (2003) presented complex visual scenes to participants, such as an office or a kitchen, and later questioned them about different aspects of the scene. They also systematically varied the presentation times of the scenes between 1 second and 10 seconds, which limited the amount of information the participants could extract from the scene. When recalling the gist of the scene, maximal performance occurred after just 1 second of presentation time. Performance for absolute location recall was above chance after a 2 second display. Colour recall performance plateaued after a 4 second display duration. Recall performance for presence, shape and relative location continued to rise as a function of display duration, beyond a display during of 10 seconds. They concluded that the extent and rate of extraction varied between the different types of information, which suggests that the system has different priorities for the processing of the different types of information they tested. There is also a vast amount of evidence suggesting that spatial location (Relative or absolute?) has a special place in the encoding of visual memory. Participants are more likely to report a stimulus that is near a target than one that is far away (Hoffman & Nelson, 1981). Selection by location is more efficient than selection by object features. (Bongartz & Sheerer, 1976; Sperling 1960, 1963; Von Wright, 1968, 1970, 1972). Participants are more accurate in detecting a stimulus, and in performing discrimination tasks concerning brightness, orientation, and form, when the target stimulus occurs at or near a cued location relative to a location which is father away (Bashinski & Bacharach, 1980; Downing 1988). Location processing occurs even when the location is not task relevant (Cave& Pashler, 1995; Cave & Zimmerman, 1997). Johnston & Pashler (1990) showed participants brief stimulus delays of different coloured letters. Participants searched for a target defined by either colour or form (the identity response), and then indicated the location of the target (the location response). Accuracy was the dependant measure. There was no evidence of identity perception without location perception. Concluded that whereas identity perception depended on location perception, location perception did not necessarily depend on identity perception. '' '' ''The prioritisation of different colours. This is a quote from Hedge (2002): “For brief stimulus durations at a given spatial frequency, visual perception is monochromatic, i.e. we perceive the stimulus as grey. At moderate presentation durations, the perception is dichromatic, i.e. we fail to perceive blue-yellow contrast variations. At longer durations, the perception is trichromatic. However the colour sensitivity co-varies with that of space and time, so that in all three cases, human subjects are maximally sensitive to medium frequencies (Wandell, 1995).” Unfortunately they do not cite the original source, but instead, a textbook on colour vision in general. If I can find the original source, we might be able to find out the time course for encoding different colours, or if certain colours are prioritised over others.